MMS 300M AntiDeepfake
Bit-exact copy of nii-yamagishilab/mms-300m-anti-deepfake (the AntiDeepfake collection, nii-yamagishilab), hosted for the Speech Anti-Spoofing Arena. Trained with RawBoost data augmentation (default).
Fonte do modelo
- Tipo
- Other
- Tamanho do arquivo
- 1,18 GB
- Modelo base
- nii-yamagishilab/mms-300m-anti-deepfake
- Versão
- vRevision 1e626a1
- Atualizado
- 25 de out. de 2026
- Downloads
- 50
- Fonte
- Hugging Face
Descrição da fonte
MMS-300M-AntiDeepfake
Bit-exact copy of nii-yamagishilab/mms-300m-anti-deepfake (the AntiDeepfake
collection, nii-yamagishilab), hosted for the Speech Anti-Spoofing Arena.
Trained with RawBoost data augmentation (default).
- Paper: Post-training for Deepfake Speech Detection, arXiv:2506.21090
- Original code: https://github.com/nii-yamagishilab/AntiDeepfake
- Backbone: MMS-300M; head = AdaptiveAvgPool1d + Linear(D, 2). Score = real logit.
- License: CC BY-NC-SA 4.0 (non-commercial) — inherited from the source.
⚠️ Evaluation restriction
It is STRICTLY PROHIBITED to evaluate this model on its training / in-domain corpora: ASVspoof2019 LA, ASVspoof2021 LA/DF, ASVspoof5, CFAD, DECRO, DFADD, "Diffuse or Confuse", DiffSSD, DSD, HABLA, MLAAD, SpoofCeleb, CVoiceFake, any LibriTTS-based set (incl. SONAR, CD-ADD, LibriSeVoc), any VoxCeleb-based set, WaveFake, any FLEURS-based set, CodecFake (any variant), any AIShell3-based set, and any MLS-based set. On the Arena it is evaluated only on ArAD and In-the-Wild.
All credit to the original authors (Ge, Wang, et al., NII).
Arena results
EER% 0.92 on XMAD EER% 0.57 on PyAra EER% 7.34 on ODSS EER% 0.62 on J-SPAW_LA EER% 4.62 on DeepVoice EER% 4.83 on InTheWild EER% 6.03 on ArAD 1-SRR% 2.27 on LRLspoof EER% 1.6 on EmoFake_test 1-SRR% 11.31 on EmoSpoofTTS arena tier arena rank
Maintainer
Maintained by Kirill Borodin (SpeechAntiSpoofingBenchmarks).
- Email: [email protected]
- Telegram: @korallll_ai
Fontes
1 fonte
Fontes
1 fonteVerificado 25 de out.
Download files
Artefatos de modelo
2 artefatos
Artefatos de modelo
2 artefatosTrechos de fonte
3 trechos
Trechos de fonte
3 trechos--- license: cc-by-nc-sa-4.0 base_model: - nii-yamagishilab/mms-300m-anti-deepfake tags: - anti-spoofing - audio-deepfake-detection - speech pipeline_tag: audio-classification --- # MMS-300M-AntiDeepfake Bit-exact copy of [`nii-yamagishilab/mms-300m-anti-deepfake`](https://huggingface.co/nii-yamagishilab/mms-300m-anti-deepfake) (the AntiDeepfake collection, nii-yamagishilab), hosted for the Speech Anti-Spoofing **Arena**. Trained with RawBoost data augmentation (default). - **Paper:** Post-training for Deepfake Speech Detection, [arXiv:2506.21090](https://arxiv.org/abs/2506.21090) - **Original code:** https://github.com/nii-yamagishilab/AntiDeepfake - **Backbone:** MMS-300M; head = AdaptiveAvgPool1d + Linear(D, 2). Score = real logit. - **License:** CC BY-NC-SA 4.0 (non-commercial) — inherited from the source. ## ⚠️ Evaluation restriction It is **STRICTLY PROHIBITED** to evaluate this model on its training / in-domain corpora: ASVspoof2019 LA, ASVspoof2021 LA/DF, ASVspoof5, CFAD, DECRO, DFADD, "Diffuse or Confuse", DiffSSD, DSD, HABLA, MLAAD, SpoofCeleb, CVoiceFake, any LibriTTS-based set (incl. SONAR, CD-ADD, LibriSeVoc), any VoxCeleb-based set, WaveFake, any FLEURS-based set, CodecFake (any variant), any AIShell3-based set, and any MLS-based set. On the Arena it is evaluated **only** on ArAD and In-the-Wild. All credit to the original authors (Ge, Wang, et al., NII). ## Arena results <!-- arena-badges --> [](https://huggingface.co/spaces/SpeechAntiSpoofingBenchmarks/SpeechAntiSpoofingArena?system=mms-300m-antideepfake) [](https://huggingface.co/spaces/SpeechAntiSpoofingBenchmarks/SpeechAntiSpoofingArena?system=mms-300m-antideepfake) [](https://huggingface.co/spaces/SpeechAntiSpoofingBenchmarks/SpeechAntiSpoofingArena?system=mms-300m-antideepfake) [](https://huggingface.co/spaces/SpeechAntiSpoofingBenchmarks/SpeechAntiSpoofingArena?system=mms-300m-antideepfake) [](https://huggingface.co/spaces/Spe...
Source context: 50 downloads · 0 likes · Pipeline audio-classification · Repo SpeechAntiSpoofingBenchmarks/MMS-300M-AntiDeepfake
Source context: 17 downloads · 0 likes · Pipeline audio-classification · Repo SpeechAntiSpoofingBenchmarks/MMS-300M-AntiDeepfake
